mqtt-to-speech

A roughly written MQTT client which receives a text message and forwards it to Amazon Polly to convert it to speech if not cached already.

The output is then cached in Redis, written to disk and played aftwards.

Requirements

• Amazon AWS credentials with permission to use Amazon Polly
• MQTT message broker
• Redis (optional)

Usage

Import class or run as stand-alone application.

• prepare mqtt server + topic
• optionally configure Redis for caching
• configure required variables (see env.dist and main() function for details)
• start program (stand-alone) or execute method start_runner (class usage)
• send MQTT message with text attribute

MQTT example message

{
  "text": "Hello World!"
}